Neurocomputing
Handprinted word recognition on a NIST data set
Machine Vision and Applications
IEEE Transactions on Pattern Analysis and Machine Intelligence
Fundamentals of Uncertainty Calculi with Applications to Fuzzy Inference
Fundamentals of Uncertainty Calculi with Applications to Fuzzy Inference
Handwritten word recognition with character and inter-character neural networks
IEEE Transactions on Systems, Man, and Cybernetics, Part B: Cybernetics
Hybrid fuzzy-neural systems in handwritten word recognition
IEEE Transactions on Fuzzy Systems
A fuzzy logic system for the detection and recognition of handwritten street numbers
IEEE Transactions on Fuzzy Systems
Comparison of crisp and fuzzy character neural networks in handwritten word recognition
IEEE Transactions on Fuzzy Systems
Lexicon-Driven Handwritten Word Recognition Using Optimal Linear Combinations of Order Statistics
IEEE Transactions on Pattern Analysis and Machine Intelligence
Lexical Search Approach for Character-String Recognition
DAS '98 Selected Papers from the Third IAPR Workshop on Document Analysis Systems: Theory and Practice
Segmentation of Japanese Handwritten Characters Using Peripheral Feature Analysis
ICPR '98 Proceedings of the 14th International Conference on Pattern Recognition-Volume 2 - Volume 2
A System for Cursive Handwritten Address Recognition
ICPR '98 Proceedings of the 14th International Conference on Pattern Recognition-Volume 2 - Volume 2
The Recognition Graph - Language Independent Adaptable On-line Cursive Script Recognition
ICDAR '05 Proceedings of the Eighth International Conference on Document Analysis and Recognition
On-line handwritten digit recognition based on trajectory and velocity modeling
Pattern Recognition Letters
Data-based adaptive computation technique
International Journal of Information and Communication Technology
Estimation of fuzzy measures using covariance matrices in Gaussian mixtures
Applied Computational Intelligence and Soft Computing
| Hi-index | 4.10 |
Handwriting recognition has challenged computer scientists for years. To succeed, a computing solution must ably recognize complex character patterns and represent imprecise, commonsense knowledge about the general appearance of characters, words, and phrases. Character recognition is a classical computing problem, dating back to neural computing's infancy. One of Frank Rosenblatt's first demonstrations on the Mark I Perceptron neurocomputer in the late 1950s involved character recognition. The Perceptron was one of the first computers based on the idea of a neural network, which is a simplified computational model of neurons in a human brain. It was the first functioning neurocomputer, and it was able to recognize a fixed-font character set. As with many artificial intelligence applications, the difficulty of handwriting recognition was greatly underestimated. Significant progress was not achieved until the late 1980s and early 1990s, when many technologies converged to enable rapid increases in recognition rates for digits, characters, and words so that reliable commercial systems could be developed. Handwriting recognition problems are either online or offline. Online recognition systems use a pressure-sensitive pad that records the pen's pressure and velocity, which would be the case with, for example, a personal digital assistant. In offline recognition, the kind we are concerned with here, system input is a digital image of handwritten letters and numbers. Handwriting recognition requires tools and techniques that recognize complex character patterns and represent imprecise, commonsense knowledge about the general appearance of characters, words, and phrases. Neural networks and fuzzy logic are complementary tools for solving such problems. Neural networks, which are highly nonlinear and highly interconnected for processing imprecise information, can finely approximate complicated decision boundaries. Fuzzy set methods can represent degrees of truth or belonging. Fuzzy logic, one of several fuzzy set methods, encodes imprecise knowledge and naturally maintains multiple hypotheses that result from the uncertainty and vagueness inherent in real problems. By combining the complementary strengths of neural and fuzzy approaches into a hybrid system, we can attain increased recognition capability for solving handwriting recognition problems. This article describes the application of neural and fuzzy methods to three problems:recognition of handwritten words,recognition of numeric fields, andlocation of handwritten street numbers in address images. These problems and methods were part of research we conducted on US Postal Service data and on problems of interest to the USPS.